Loading...
Please wait, while we are loading the content...
Similar Documents
A Neural Beamspace-Domain Filter for Real-Time Multi-Channel Speech Enhancement
Content Provider | MDPI |
---|---|
Author | Liu, Wenzhe Li, Andong Wang, Xiao Yuan, Minmin Chen, Yi Zheng, Chengshi Li, Xiaodong |
Copyright Year | 2022 |
Description | Most deep-learning-based multi-channel speech enhancement methods focus on designing a set of beamforming coefficients, to directly filter the low signal-to-noise ratio signals received by microphones, which hinders the performance of these approaches. To handle these problems, this paper designs a causal neural filter that fully exploits the spectro-temporal-spatial information in the beamspace domain. Specifically, multiple beams are designed to steer towards all directions, using a parameterized super-directive beamformer in the first stage. After that, a deep-learning-based filter is learned by, simultaneously, modeling the spectro-temporal-spatial discriminability of the speech and the interference, so as to extract the desired speech, coarsely, in the second stage. Finally, to further suppress the interference components, especially at low frequencies, a residual estimation module is adopted, to refine the output of the second stage. Experimental results demonstrate that the proposed approach outperforms many state-of-the-art (SOTA) multi-channel methods, on the generated multi-channel speech dataset based on the DNS-Challenge dataset. |
Starting Page | 1081 |
e-ISSN | 20738994 |
DOI | 10.3390/sym14061081 |
Journal | Symmetry |
Issue Number | 6 |
Volume Number | 14 |
Language | English |
Publisher | MDPI |
Publisher Date | 2022-05-24 |
Access Restriction | Open |
Subject Keyword | Symmetry Artificial Intelligence Industrial Engineering Multi-channel Speech Enhancement Neural Beam Filter Deep Learning |
Content Type | Text |
Resource Type | Article |